Hugging Face's logo

qualcomm
/
GPUNet

Image Classification
PyTorch
backbone
bu_auto
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
1557118
·
verified ·
1 Parent(s): a85d943

v0.49.1

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.49.1 for changelog.

Files changed (1) hide show
  1. README.md +101 -101
README.md CHANGED
@@ -16,7 +16,7 @@ pipeline_tag: image-classification
16
  GPUNet is a machine learning model that can classify images from the Imagenet dataset. It can also be used as a backbone in building more complex models for specific use cases.
17
 
18
  This is based on the implementation of GPUNet found [here](https://github.com/NVIDIA/DeepLearningExamples/tree/master/PyTorch/Classification/GPUNet).
19
- This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/gpunet) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
20
 
21
  Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
22
 
@@ -29,28 +29,28 @@ Below are pre-exported model assets ready for deployment.
29
 
30
  | Runtime | Precision | Chipset | SDK Versions | Download |
31
  |---|---|---|---|---|
32
- | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.48.0/gpunet-onnx-float.zip)
33
- | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.48.0/gpunet-onnx-w8a16.zip)
34
- | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.48.0/gpunet-onnx-w8a8.zip)
35
- | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.48.0/gpunet-qnn_dlc-float.zip)
36
- | QNN_DLC | w8a16 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.48.0/gpunet-qnn_dlc-w8a16.zip)
37
- | QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.48.0/gpunet-qnn_dlc-w8a8.zip)
38
- | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.48.0/gpunet-tflite-float.zip)
39
- | TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.48.0/gpunet-tflite-w8a8.zip)
40
 
41
  For more device-specific assets and performance metrics, visit **[GPUNet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/gpunet)**.
42
 
43
 
44
  ### Option 2: Export with Custom Configurations
45
 
46
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/gpunet) Python library to compile and export the model with your own:
47
  - Custom weights (e.g., fine-tuned checkpoints)
48
  - Custom input shapes
49
  - Target device and runtime configurations
50
 
51
  This option is ideal if you need to customize the model beyond the default configuration provided here.
52
 
53
- See our repository for [GPUNet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/gpunet) for usage instructions.
54
 
55
  ## Model Details
56
 
@@ -66,98 +66,98 @@ See our repository for [GPUNet on GitHub](https://github.com/qualcomm/ai-hub-mod
66
  ## Performance Summary
67
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
68
  |---|---|---|---|---|---|---
69
- | GPUNet | ONNX | float | Snapdragon® X2 Elite | 0.468 ms | 24 - 24 MB | NPU
70
- | GPUNet | ONNX | float | Snapdragon® X Elite | 1.118 ms | 24 - 24 MB | NPU
71
- | GPUNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.786 ms | 0 - 61 MB | NPU
72
- | GPUNet | ONNX | float | Qualcomm® QCS8550 (Proxy) | 1.042 ms | 1 - 2 MB | NPU
73
- | GPUNet | ONNX | float | Qualcomm® QCS9075 | 1.358 ms | 1 - 3 MB | NPU
74
- | GPUNet | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.609 ms | 0 - 37 MB | NPU
75
- | GPUNet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.508 ms | 0 - 36 MB | NPU
76
- | GPUNet | ONNX | w8a16 | Snapdragon® X2 Elite | 0.393 ms | 12 - 12 MB | NPU
77
- | GPUNet | ONNX | w8a16 | Snapdragon® X Elite | 0.999 ms | 12 - 12 MB | NPU
78
- | GPUNet | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.58 ms | 0 - 65 MB | NPU
79
- | GPUNet | ONNX | w8a16 | Qualcomm® QCS6490 | 97.936 ms | 23 - 33 MB | CPU
80
- | GPUNet | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 0.825 ms | 0 - 16 MB | NPU
81
- | GPUNet | ONNX | w8a16 | Qualcomm® QCS9075 | 0.971 ms | 0 - 3 MB | NPU
82
- | GPUNet | ONNX | w8a16 | Qualcomm® QCM6690 | 49.794 ms | 28 - 35 MB | CPU
83
- | GPUNet | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.434 ms | 0 - 45 MB | NPU
84
- | GPUNet | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 38.154 ms | 30 - 38 MB | CPU
85
- | GPUNet | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.36 ms | 0 - 49 MB | NPU
86
- | GPUNet | ONNX | w8a8 | Snapdragon® X2 Elite | 0.309 ms | 12 - 12 MB | NPU
87
- | GPUNet | ONNX | w8a8 | Snapdragon® X Elite | 0.729 ms | 12 - 12 MB | NPU
88
- | GPUNet | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.472 ms | 0 - 62 MB | NPU
89
- | GPUNet | ONNX | w8a8 | Qualcomm® QCS6490 | 16.511 ms | 0 - 13 MB | CPU
90
- | GPUNet | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.614 ms | 0 - 2 MB | NPU
91
- | GPUNet | ONNX | w8a8 | Qualcomm® QCS9075 | 0.723 ms | 0 - 3 MB | NPU
92
- | GPUNet | ONNX | w8a8 | Qualcomm® QCM6690 | 10.212 ms | 6 - 14 MB | CPU
93
- | GPUNet | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.392 ms | 0 - 47 MB | NPU
94
- | GPUNet | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 7.805 ms | 6 - 15 MB | CPU
95
- | GPUNet | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.362 ms | 0 - 48 MB | NPU
96
- | GPUNet | QNN_DLC | float | Snapdragon® X2 Elite | 0.646 ms | 1 - 1 MB | NPU
97
- | GPUNet | QNN_DLC | float | Snapdragon® X Elite | 1.414 ms | 1 - 1 MB | NPU
98
- | GPUNet | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.915 ms | 0 - 62 MB | NPU
99
- | GPUNet | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 4.664 ms | 1 - 33 MB | NPU
100
- | GPUNet | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 1.286 ms | 1 - 136 MB | NPU
101
- | GPUNet | QNN_DLC | float | Qualcomm® SA8775P | 1.756 ms | 1 - 36 MB | NPU
102
- | GPUNet | QNN_DLC | float | Qualcomm® QCS9075 | 1.595 ms | 3 - 5 MB | NPU
103
- | GPUNet | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 2.404 ms | 0 - 65 MB | NPU
104
- | GPUNet | QNN_DLC | float | Qualcomm® SA7255P | 4.664 ms | 1 - 33 MB | NPU
105
- | GPUNet | QNN_DLC | float | Qualcomm® SA8295P | 2.305 ms | 0 - 33 MB | NPU
106
- | GPUNet | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.706 ms | 0 - 33 MB | NPU
107
- | GPUNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.572 ms | 1 - 37 MB | NPU
108
- | GPUNet | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 0.564 ms | 0 - 0 MB | NPU
109
- | GPUNet | QNN_DLC | w8a16 | Snapdragon® X Elite | 1.261 ms | 0 - 0 MB | NPU
110
- | GPUNet | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.75 ms | 0 - 58 MB | NPU
111
- | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 3.206 ms | 0 - 2 MB | NPU
112
- | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 2.443 ms | 0 - 42 MB | NPU
113
- | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 1.074 ms | 0 - 83 MB | NPU
114
- | GPUNet | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.29 ms | 0 - 44 MB | NPU
 
115
  | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 1.226 ms | 0 - 2 MB | NPU
116
- | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 6.565 ms | 0 - 159 MB | NPU
117
- | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 1.415 ms | 0 - 61 MB | NPU
118
- | GPUNet | QNN_DLC | w8a16 | Qualcomm® SA7255P | 2.443 ms | 0 - 42 MB | NPU
119
- | GPUNet | QNN_DLC | w8a16 | Qualcomm® SA8295P | 1.707 ms | 0 - 39 MB | NPU
120
- | GPUNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.534 ms | 0 - 44 MB | NPU
121
- | GPUNet | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.28 ms | 0 - 42 MB | NPU
122
- | GPUNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.429 ms | 0 - 45 MB | NPU
123
- | GPUNet | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.37 ms | 0 - 0 MB | NPU
124
- | GPUNet | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.722 ms | 0 - 0 MB | NPU
125
- | GPUNet | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.466 ms | 0 - 56 MB | NPU
126
- | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 1.998 ms | 2 - 4 MB | NPU
127
- | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 1.4 ms | 0 - 38 MB | NPU
128
- | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.606 ms | 0 - 2 MB | NPU
129
- | GPUNet | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.81 ms | 0 - 41 MB | NPU
130
- | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.703 ms | 0 - 2 MB | NPU
131
- | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 3.523 ms | 0 - 41 MB | NPU
132
- | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.848 ms | 0 - 58 MB | NPU
133
- | GPUNet | QNN_DLC | w8a8 | Qualcomm® SA7255P | 1.4 ms | 0 - 38 MB | NPU
134
- | GPUNet | QNN_DLC | w8a8 | Qualcomm® SA8295P | 1.077 ms | 0 - 36 MB | NPU
135
- | GPUNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.332 ms | 0 - 37 MB | NPU
136
  | GPUNet | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.796 ms | 0 - 38 MB | NPU
137
- | GPUNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.292 ms | 0 - 44 MB | NPU
138
- | GPUNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.905 ms | 16 - 112 MB | NPU
139
- | GPUNet | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 4.638 ms | 0 - 62 MB | NPU
140
- | GPUNet | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 1.291 ms | 0 - 151 MB | NPU
141
- | GPUNet | TFLITE | float | Qualcomm® SA8775P | 1.751 ms | 0 - 64 MB | NPU
142
- | GPUNet | TFLITE | float | Qualcomm® QCS9075 | 1.581 ms | 0 - 27 MB | NPU
143
- | GPUNet | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 2.403 ms | 0 - 94 MB | NPU
144
- | GPUNet | TFLITE | float | Qualcomm® SA7255P | 4.638 ms | 0 - 62 MB | NPU
145
- | GPUNet | TFLITE | float | Qualcomm® SA8295P | 2.291 ms | 0 - 61 MB | NPU
146
- | GPUNet | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.703 ms | 0 - 62 MB | NPU
147
- | GPUNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.564 ms | 0 - 66 MB | NPU
148
- | GPUNet | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.333 ms | 0 - 54 MB | NPU
149
- | GPUNet | TFLITE | w8a8 | Qualcomm® QCS6490 | 1.583 ms | 0 - 15 MB | NPU
150
- | GPUNet | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 1.101 ms | 0 - 37 MB | NPU
151
- | GPUNet | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.434 ms | 0 - 2 MB | NPU
152
- | GPUNet | TFLITE | w8a8 | Qualcomm® SA8775P | 0.641 ms | 0 - 41 MB | NPU
153
- | GPUNet | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.528 ms | 0 - 14 MB | NPU
154
- | GPUNet | TFLITE | w8a8 | Qualcomm® QCM6690 | 3.044 ms | 0 - 39 MB | NPU
155
- | GPUNet | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.674 ms | 0 - 56 MB | NPU
156
- | GPUNet | TFLITE | w8a8 | Qualcomm® SA7255P | 1.101 ms | 0 - 37 MB | NPU
157
- | GPUNet | TFLITE | w8a8 | Qualcomm® SA8295P | 0.906 ms | 0 - 35 MB | NPU
158
- | GPUNet | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.263 ms | 0 - 35 MB | NPU
159
- | GPUNet | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.616 ms | 0 - 37 MB | NPU
160
- | GPUNet | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.234 ms | 0 - 43 MB | NPU
161
 
162
  ## License
163
  * The license for the original implementation of GPUNet can be found
 
16
  GPUNet is a machine learning model that can classify images from the Imagenet dataset. It can also be used as a backbone in building more complex models for specific use cases.
17
 
18
  This is based on the implementation of GPUNet found [here](https://github.com/NVIDIA/DeepLearningExamples/tree/master/PyTorch/Classification/GPUNet).
19
+ This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/gpunet) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
20
 
21
  Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
22
 
 
29
 
30
  | Runtime | Precision | Chipset | SDK Versions | Download |
31
  |---|---|---|---|---|
32
+ | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.49.1/gpunet-onnx-float.zip)
33
+ | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.49.1/gpunet-onnx-w8a16.zip)
34
+ | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.49.1/gpunet-onnx-w8a8.zip)
35
+ | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.49.1/gpunet-qnn_dlc-float.zip)
36
+ | QNN_DLC | w8a16 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.49.1/gpunet-qnn_dlc-w8a16.zip)
37
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.49.1/gpunet-qnn_dlc-w8a8.zip)
38
+ | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.49.1/gpunet-tflite-float.zip)
39
+ | TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gpunet/releases/v0.49.1/gpunet-tflite-w8a8.zip)
40
 
41
  For more device-specific assets and performance metrics, visit **[GPUNet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/gpunet)**.
42
 
43
 
44
  ### Option 2: Export with Custom Configurations
45
 
46
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/gpunet) Python library to compile and export the model with your own:
47
  - Custom weights (e.g., fine-tuned checkpoints)
48
  - Custom input shapes
49
  - Target device and runtime configurations
50
 
51
  This option is ideal if you need to customize the model beyond the default configuration provided here.
52
 
53
+ See our repository for [GPUNet on GitHub](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/gpunet) for usage instructions.
54
 
55
  ## Model Details
56
 
 
66
  ## Performance Summary
67
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
68
  |---|---|---|---|---|---|---
69
+ | GPUNet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.507 ms | 0 - 36 MB | NPU
70
+ | GPUNet | ONNX | float | Snapdragon® X2 Elite | 0.477 ms | 24 - 24 MB | NPU
71
+ | GPUNet | ONNX | float | Snapdragon® X Elite | 1.138 ms | 24 - 24 MB | NPU
72
+ | GPUNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.78 ms | 0 - 59 MB | NPU
73
+ | GPUNet | ONNX | float | Qualcomm® QCS8550 (Proxy) | 1.045 ms | 1 - 3 MB | NPU
74
+ | GPUNet | ONNX | float | Qualcomm® QCS9075 | 1.349 ms | 1 - 3 MB | NPU
75
+ | GPUNet | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.601 ms | 0 - 30 MB | NPU
76
+ | GPUNet | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.351 ms | 0 - 49 MB | NPU
77
+ | GPUNet | ONNX | w8a16 | Snapdragon® X2 Elite | 0.388 ms | 12 - 12 MB | NPU
78
+ | GPUNet | ONNX | w8a16 | Snapdragon® X Elite | 0.974 ms | 12 - 12 MB | NPU
79
+ | GPUNet | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.589 ms | 0 - 61 MB | NPU
80
+ | GPUNet | ONNX | w8a16 | Qualcomm® QCS6490 | 100.92 ms | 22 - 32 MB | CPU
81
+ | GPUNet | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 0.821 ms | 0 - 3 MB | NPU
82
+ | GPUNet | ONNX | w8a16 | Qualcomm® QCS9075 | 0.977 ms | 0 - 3 MB | NPU
83
+ | GPUNet | ONNX | w8a16 | Qualcomm® QCM6690 | 49.458 ms | 29 - 37 MB | CPU
84
+ | GPUNet | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.43 ms | 0 - 42 MB | NPU
85
+ | GPUNet | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 38.092 ms | 29 - 38 MB | CPU
86
+ | GPUNet | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.354 ms | 0 - 48 MB | NPU
87
+ | GPUNet | ONNX | w8a8 | Snapdragon® X2 Elite | 0.312 ms | 12 - 12 MB | NPU
88
+ | GPUNet | ONNX | w8a8 | Snapdragon® X Elite | 0.722 ms | 12 - 12 MB | NPU
89
+ | GPUNet | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.473 ms | 0 - 61 MB | NPU
90
+ | GPUNet | ONNX | w8a8 | Qualcomm® QCS6490 | 16.265 ms | 1 - 14 MB | CPU
91
+ | GPUNet | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.616 ms | 0 - 2 MB | NPU
92
+ | GPUNet | ONNX | w8a8 | Qualcomm® QCS9075 | 0.718 ms | 0 - 3 MB | NPU
93
+ | GPUNet | ONNX | w8a8 | Qualcomm® QCM6690 | 10.304 ms | 6 - 15 MB | CPU
94
+ | GPUNet | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.393 ms | 0 - 39 MB | NPU
95
+ | GPUNet | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 7.748 ms | 7 - 16 MB | CPU
96
+ | GPUNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.569 ms | 1 - 37 MB | NPU
97
+ | GPUNet | QNN_DLC | float | Snapdragon® X2 Elite | 0.651 ms | 1 - 1 MB | NPU
98
+ | GPUNet | QNN_DLC | float | Snapdragon® X Elite | 1.437 ms | 1 - 1 MB | NPU
99
+ | GPUNet | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.922 ms | 0 - 63 MB | NPU
100
+ | GPUNet | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 4.701 ms | 1 - 33 MB | NPU
101
+ | GPUNet | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 1.29 ms | 1 - 2 MB | NPU
102
+ | GPUNet | QNN_DLC | float | Qualcomm® SA8775P | 1.757 ms | 1 - 36 MB | NPU
103
+ | GPUNet | QNN_DLC | float | Qualcomm® QCS9075 | 1.574 ms | 1 - 3 MB | NPU
104
+ | GPUNet | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 2.412 ms | 0 - 61 MB | NPU
105
+ | GPUNet | QNN_DLC | float | Qualcomm® SA7255P | 4.701 ms | 1 - 33 MB | NPU
106
+ | GPUNet | QNN_DLC | float | Qualcomm® SA8295P | 2.299 ms | 1 - 34 MB | NPU
107
+ | GPUNet | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.713 ms | 0 - 36 MB | NPU
108
+ | GPUNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.44 ms | 0 - 44 MB | NPU
109
+ | GPUNet | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 0.569 ms | 0 - 0 MB | NPU
110
+ | GPUNet | QNN_DLC | w8a16 | Snapdragon® X Elite | 1.235 ms | 0 - 0 MB | NPU
111
+ | GPUNet | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.762 ms | 0 - 58 MB | NPU
112
+ | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 3.226 ms | 0 - 2 MB | NPU
113
+ | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 2.462 ms | 0 - 42 MB | NPU
114
+ | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 1.062 ms | 0 - 2 MB | NPU
115
+ | GPUNet | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.295 ms | 0 - 44 MB | NPU
116
  | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 1.226 ms | 0 - 2 MB | NPU
117
+ | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 6.537 ms | 0 - 159 MB | NPU
118
+ | GPUNet | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 1.435 ms | 0 - 60 MB | NPU
119
+ | GPUNet | QNN_DLC | w8a16 | Qualcomm® SA7255P | 2.462 ms | 0 - 42 MB | NPU
120
+ | GPUNet | QNN_DLC | w8a16 | Qualcomm® SA8295P | 1.715 ms | 0 - 38 MB | NPU
121
+ | GPUNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.524 ms | 0 - 40 MB | NPU
122
+ | GPUNet | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.274 ms | 0 - 42 MB | NPU
123
+ | GPUNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.288 ms | 0 - 44 MB | NPU
124
+ | GPUNet | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.363 ms | 0 - 0 MB | NPU
125
+ | GPUNet | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.73 ms | 0 - 0 MB | NPU
126
+ | GPUNet | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.458 ms | 0 - 55 MB | NPU
127
+ | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 2.001 ms | 2 - 4 MB | NPU
128
+ | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 1.407 ms | 0 - 38 MB | NPU
129
+ | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.612 ms | 0 - 2 MB | NPU
130
+ | GPUNet | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.814 ms | 0 - 42 MB | NPU
131
+ | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.701 ms | 2 - 4 MB | NPU
132
+ | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 3.53 ms | 0 - 41 MB | NPU
133
+ | GPUNet | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.844 ms | 0 - 58 MB | NPU
134
+ | GPUNet | QNN_DLC | w8a8 | Qualcomm® SA7255P | 1.407 ms | 0 - 38 MB | NPU
135
+ | GPUNet | QNN_DLC | w8a8 | Qualcomm® SA8295P | 1.121 ms | 0 - 36 MB | NPU
136
+ | GPUNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.346 ms | 0 - 40 MB | NPU
137
  | GPUNet | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.796 ms | 0 - 38 MB | NPU
138
+ | GPUNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.566 ms | 0 - 67 MB | NPU
139
+ | GPUNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.899 ms | 0 - 94 MB | NPU
140
+ | GPUNet | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 4.632 ms | 0 - 62 MB | NPU
141
+ | GPUNet | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 1.279 ms | 0 - 2 MB | NPU
142
+ | GPUNet | TFLITE | float | Qualcomm® SA8775P | 1.747 ms | 0 - 65 MB | NPU
143
+ | GPUNet | TFLITE | float | Qualcomm® QCS9075 | 1.58 ms | 0 - 27 MB | NPU
144
+ | GPUNet | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 2.402 ms | 0 - 94 MB | NPU
145
+ | GPUNet | TFLITE | float | Qualcomm® SA7255P | 4.632 ms | 0 - 62 MB | NPU
146
+ | GPUNet | TFLITE | float | Qualcomm® SA8295P | 2.26 ms | 0 - 61 MB | NPU
147
+ | GPUNet | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.704 ms | 0 - 62 MB | NPU
148
+ | GPUNet | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.233 ms | 0 - 43 MB | NPU
149
+ | GPUNet | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.343 ms | 0 - 54 MB | NPU
150
+ | GPUNet | TFLITE | w8a8 | Qualcomm® QCS6490 | 1.639 ms | 0 - 15 MB | NPU
151
+ | GPUNet | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 1.112 ms | 0 - 37 MB | NPU
152
+ | GPUNet | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.424 ms | 0 - 79 MB | NPU
153
+ | GPUNet | TFLITE | w8a8 | Qualcomm® SA8775P | 0.641 ms | 0 - 40 MB | NPU
154
+ | GPUNet | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.529 ms | 0 - 14 MB | NPU
155
+ | GPUNet | TFLITE | w8a8 | Qualcomm® QCM6690 | 3.047 ms | 0 - 38 MB | NPU
156
+ | GPUNet | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.686 ms | 0 - 56 MB | NPU
157
+ | GPUNet | TFLITE | w8a8 | Qualcomm® SA7255P | 1.112 ms | 0 - 37 MB | NPU
158
+ | GPUNet | TFLITE | w8a8 | Qualcomm® SA8295P | 0.93 ms | 0 - 35 MB | NPU
159
+ | GPUNet | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.273 ms | 0 - 40 MB | NPU
160
+ | GPUNet | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.615 ms | 0 - 37 MB | NPU
 
161
 
162
  ## License
163
  * The license for the original implementation of GPUNet can be found